- Email: [email protected]
BHK cell variant with defective fibronectin receptor function
Cell Biology
International
Reports,
Vol. 8, No. 2, February
171
1984
BHK CELL VARIANT WITH DEFECTIVE FIBRONECTIN RECEPTOR FUNCTION Nancy Oppenheimer-Marks, Barbara Border, and Frederick Grinnell Department of Cell Biology University of Texas Health Science Center Dallas, Texas 75235 Running
title:
Fibronectin-receptor
defective
BHK cell
variants
Baby hamster kidney cells were mutagenized with ABSTRACT. N-methyl-N'-nitro-N-nitrosoguanidine and selected to obtain a The cell variant FN-1 was population of non-attaching cells. recloned, and cloned from the non-attaching cell population, tested for cell adhesive interactions using four different assays of fibronectin (pFN) receptor function: cell attachment and spreading on culture dishes and cell binding and phagocytosis of latex beads. On pFN-coated culture dishes, FN-1 cells had decreased attachment compared to parental cells and were unable to as many pFN-bead spread. With pFN-coated beads, only one third binding sites could be detected on FN-1 cells as on the parental cells, and the FN-1 cells were unable to phagocytose the pFNIn other studies, the variant cells were able to coated beads. attach normally and spread partially on substrata coated with polycationic ferritin, concanavalin A, or anti-BHK cell surface antibody. The results suggest that the pFN-receptor function of FN-1 cells is defective. The attachment and spreading of fibroblasts has INTRODUCTION. been shown to occur on a variety of naturally occurring (e.g., fibronectin) or experimental (e.g., lectins) ligand-coated substrata (reviewed in Grinnell, 1983). The meaning of the redundancy in ligands that can promote cell adhesion is unclear. One possibility is that there are independent receptors on the cell surface for each ligand, and that binding between receptor and ligand gives rise to a common signal resulting in cytoskeletal reorganization. Such a model is analagous to that proposed for capping of independent ligands on the cell surface (Raff and DePetris, 1973; Singer -et -', al 1978). One approach to studying this problem has been the use of cell variants with altered adhesive properties. Harper and Juliano (1980), have described a CHO variant that was unable to attach or spread on pFN-coated substrata but attached and spread normally on substrata coated with extracellular matrix, Con A, or poly-L-lysine. Also, Briles and Haskew (1982) described a hepatoma cell variant that attached and spread normally on 0309-l
651/841020171-08/$03.00/O
@ 1984 Academic
Press Inc. (London)
Ltd.
172
Cell Biology
international
Reports,
Vol. 8, No. 2, February
1984
pFN-coated substrata but not on collagen films. One question unresolved by these studies, however, was whether the cell surface receptors were still functional on the cells that failed to attach to the ligand-coated substrata. For instance, Juliano has found that the CHO cell variant incapable of attaching on pFN-coated substrata (see above) binds pFN-coated beads as well as parental cells (Cheung and Juliano, 1983). In order to learn more about the pFN receptor we have begun to study adhesion-deficient BHK cell variants. In these studies, we have used four different assays of pFN-receptor function: cell-substratum attachment, cell-substratum spreading, cell-bead binding, and cell-bead phagocytosis. In addition, the ability of the variant cells to attach and spread on a variety of different ligand-coated substrata has been characterized. The results of these studies are reported herein. MATERIALS AND METHODS. Cells. BHK cells were obtained from Dr. Adrian Chappel (Center for Disease Control, Atlanta, GA) and grown in suspension culture (Grinnell et al., 1980). Exponentially growing cells were harvested bycezrifugation at 800 x g (Sorvall HL-4 rotor) for 4 minutes at room temperature. Cells were washed in Dulbecco's PBS (DPBS) (150 mM NaCl, 3 mM KCl, 6 mM NazHPO,, 1 mM KHzP04, 1 mM CaClz, 0.5 mM MgCl,, pH 7.2) containing 1% bovine serum albumin (BSA). Preparation of FN-1 Cells. BHK cells (107) in 40 ml of minimal essential medium (MEM) were mutagenized with N-methyl-N'-nitro-Nnitrosoguanidine (MNNG) (1 pg/ml) for 4 hours at 37O. The cells were removed from the mutagen by centrifugation for 3 min at 800 x in 20 ml of MEM plus 10% fetal calf serum, and g, resuspended incubated at 37' for 7 days to allow for recovery from the mutagen. Mutagenized BHK cells were subjected to selection by an enrichment procedure to obtain a population of non-attaching cells. On day 1, mutagenized cells (3 x lo7 in 150 ml MEM containing 10% serum) were incubated in Falcon 250 ml tissue attaching culture flasks for 10 min at 37O to remove rapidly were collected, centrifuged, cells. Loosely attached cells resuspended in 100 ml of MEM plus 10% serum and incubated for 2 hours in 250 ml tissue culture flasks at 37O. This procedure was repeated twice and by the end of the third cycle 4.6% of the starting mutagenized cell population had been selected as nonattaching cells. The non-attached cells then were grown in At this point, the suspension culture to obtain 14 x lo7 cells. selection procedure, three 2 hour incubation cycles in tissue Selection was carried out a total culture flasks, was repeated. of 15 times (3 attachment cycles each time) until 95% of the cell population was unable to attach to tissue culture flasks in MEM
Cell Biology
plus
International
Reports,
Vol. 8, No. 2, February
1984
173
10% serum.
After the last selection cycle, the non-attached cells were Cells were diluted in MEM plus 10% cloned by limiting dilution. Aliquots (1 ml) serum to a final concentration of 0.5 cells/ml. of the diluted cells then were incubated in 24 well Falcon tissue Clones were incubator. culture dishes at 37O in a humidified for their ability to grown to high cell density and tested The variant line, FN-1, which interact with pFN-coated substrata. is described in this paper, was recloned. Human plasma fibronectin (pFN) Fibronectin and other materials. was prepared by ammonium sulfate precipitation and ion-exchange The specific activity of chromatography (Grinnell et al., 1980). -the plasma fibronectin preparations used was ZOO-400 units/mg. One unit of activity has been defined as the amount of fibronectin required to promote complete cell spreading in a standardized Radiolabeled 3H-pFN was prepared assay (Grinnell al 1980). -et -** similarly as described previously (Grinnell, 1980). The specific activity of radiolabeled fibronectin resulting from this procedure was 9,000-16,000 cpm/pg. Crystalline bovine serum albumin (BSA) and polycationic Concanavaferritin (PCF) were obtained from Miles Laboratories. from Sigma Chemical Co. lin A (Con A), and MNNG were purchased Polystyrene latex beads with a diameter of 0.76 uM were obtained from Dow Diagnostics. Minimal essential medium, 0.25% trypsin and fetal bovine serum were obtained from GIBCo. solution, Preparation of ligand-coated beads and substrata. Polystyrene latex beads were coated with 'H-pFN as described previously 1980). (Grinnell, The conditions were adjusted to give a final pFN concentration on the beads of 350 ng/cm2 or greater. Just prior to adding the beads to the cells, the bead preparations were sonicated for 5 seconds at 60 watts (Heat Systems Ultrasonics, Inc.) in order to disperse bead aggregates. Falcon #3001 (35 mm) tissue culture dishes were coated with pFN or other ligands as described previously (Grinnell and Hays, 1978) at the ligand concentrations indicated. Dishes coated with ligands were countercoated with DPBS - 1% heat denatured BSA and rinsed with DPBS immediately prior to adding the cells (Grinnell and Feld, 1979). Adhesion and bead binding assays. Cell attachment and cell spreading were measured using standard techniques (Grinnell et al., 1980). To measure bead binding and phagocytosis, cells (2; i?) were mixed with 3H pFN-coated 0.76 uM beads as indicated in 1.0 ml DPBS - 1% BSA. After 20 min at the temperature indicated, the samples were washed, resuspended in 1.0 ml of DPBS - 1% BSA and centrifuged through 2.0 ml of DPBS - 5% BSA. Unbound beads
174
Cell Biology
International
Reports,
Vol. 8, No. 2, February
1984
100
I- 80 E P ” 60 i! s j
40
Y
.-m.
s
FN-1
/
20
0 LA 0
50
loo
Figure
1.
Cell
PCF img/mll
Con A ~pglmll
pFN tpg/ml)
attachment
to various
AMI-EHK
(j.tglmll
substrata
Substrata were coated with the concentrations of the indicated. Coated substrata were incubated with parental cells and the extent of cell attachment measured. Other are in Materials and Methods.
--
8
A .O
PARENT
C -0-o
-PO
ligands or FN-1 details
D
0
0
Figure
2.
Cell
FN-I w..-0 50 1b6 pFN lpglml)
spreading
0
50
100500
Con A fpg/ml)
on various
0
1.0 2.0 0 50 100 200 PCF (mglml) Anti-BHK (pglml)
substrata
Substrata were coated with the concentrations of the ligands indicated. Coated substrata were incubated with parental or FN-1 cells and the extent of cell spreading measured. Other details are in Materials and Methods.
Cell Biology
International
Reports,
Vol. 8, No. 2, February
1984
with cells and whereas cells remained in the supernatant; the sedimented cells were Samples of sedimented. bound beads mixed with 10 ml of Budget Solve (RPI Corp.), and radioactivity was measured in a Nuclear Chicago Mark II scintillation spectroThe remaining sedimented cells were mixed with 1.0 ml photometer. of 0.25% trypsin and incubated for 5 min at 37O to remove surface bound beads but not ingested beads. After these incubations were stopped by the addition of 1.0 ml of fetal calf serum, the cells resuspended in 1.0 ml of DPBS - 1% BSA, and were centrifuged, Beads that were centrifuged through 2 ml of DPBS - 5% BSA. removed by trypsin treatment remained in the supernatant; whereas Samples of the cells and cells with ingested beads sedimented. The sedimented cells were analysed for radioactivity as above. number of total beads bound per cell was calculated as: (cell bead)(cells/ radioactivity) f (radioactivity/ associated The number of ingested beads per cell was calculated incubation). similarly using the cell ingested radioactivity (see McAbee and Grinnell, 1983). RESULTS. BHX cells were mutagenCell Adhesion to Ligand-Coated Substrata. in serum-containing ized with MNNG. Cells that were non-adhesive medium were selected and cloned as described in Materials and The variant clone "FN-1" did not attach to pFN-coated Methods. substrata pre-coated with 5 ug/ml of pFN, which was a concentration sufficient to permit about 90% of the parental cells to As the pFN-concentration was increased, some attach (Figure 1A). attachment of FN-1 cells was observed with 28% of the cells attaching at pFN concentrations of 20 ug/ml or higher. At all pFN concentrations tested, however, FN-1 cells were unable to spread (Figure 2A). When tested on substrata coated with different experimental ligands, FN-1 cells were found to attach to Con A, polycationic ferritin (PCF), and antibodies to BHK cell surfaces (anti-BRK) almost as well as parental cells (Figure lB-D). Spreading of the variant cells occurred on these substrata, but was less than that observed with the parentals (Figure 2B-D). In many cases, the variant cells appeared to extend filipodia but were unable to complete the spreading process. Cell Binding of pFN-Coated Beads. The inability of cells to attach and spread on pFN-coated substrata is not sufficient evidence that pFN receptors are missing or defective (see Introduction). Therefore, experiments also were carried out to determine how well FN-1 cells interacted with pFN-coated beads. Previous studies have demonstrated that the cell-bead binding assay has characteristics typical of a direct ligand-receptor binding assay (Grinnell, 1980; McAbee and Grinnell, 1983). As shown in Table I, the ability of FN-1 cells to bind pFN-coated beads was less than that of the parental cells. On average, the
Cell Biology
176
Table
I.
Binding
International
of pFN-coated
Bead Number
Reports,
Vol. 8, No. 2, February
beads to cells
at 40a
Beads Bound Per Cellb
Added
Parental
% Binding
FN-1 Cells
Cells
FN-l/Parent
3.1 x 109
47 2 2.1
16.5 2 2.1
35.1
6.25
71.2
27.0
5 1.79
37.9
49.7
2 1.16
34.1
x 10'
1.25 x lOlo
145.8
5 18.7
1984
Parental or FN-1 cells were incubated with 3H-pFN-coated a) beads for 20 min at 4O. At the end of the incubations, the radioactivity associated with the cells was measured, and bead binding calculated. Other details are in Materials and Methods.
b)
Mean and S.D.
Table
II.
of duplicate
Phagocytosis
experiments.
of pFN-coated
Beads
beads to cells
Beads Ingested
Cell
Total
Type
Bound/Cellb
Parent
158 + 20.8
50.0
FN-1
40.7
2.1 k .21
2 .49
per Cellb
a>
k 1.16
at 370a
% Ingestion 31.6 5.1
Parental or FN-1 cells were incubated with 2.1 x 10' 3H-pFNcoated beads for 20 min at 37O. At the end of the incubations, the total radioactivity associated with the cells and the ingested radioactivity were measured, and bead binding and ingestion were calculated. Other details are in Materials and Methods.
b)
Mean and S.D.
of duplicate
experiments
Cell Biology
international
variants appeared the parentals.
Reports,
to have
Vol. 8, No. 2, February
35% of the
receptor
1984
binding
capacity
of
The ability of FN-1 cells to phagocytose pFN-coated beads As shown in Table II, total bead binding at 37O also was tested. (surface bound plus ingested) by the variant cells was about 25% Of the beads bound to the parental of that of the parental cells. 31.6% were inside the cells; whereas, only 5.1% of the cells, It can be total bound beads were inside the variant cells. concluded, therefore, that the variants are deficient not only in spreading on pFN-coated substrata, but also, in phagocytosis of pFN-coated beads. The purpose of the studies described in this paper DISCUSSION. was to use adhesion-defective BHE cells to learn more about the involved in cell attachment and spreading. The adhesion sites variant described, FN-1, was shown to have decreased cell attachment and no cell spreading on pFN-coated substrata. Also, of cell surface-pFN bead binding activity was only one third this is the strongest evidence for a detected. To our knowledge, cell variant with inactive pFN receptors. Also, this is the first instance in which a phagocytosis-deficient cell variant has been described. Previously, the possibility was suggested that cell spreading and phagocytosis involve similar cell responses to different sized substrata (Grinnell, 1980). The inability of FN-1 cells to phagocytose pFN-coated beads provides further evidence that spreading and phagocytosis are closely related. At the present time, studies have been initiated to try to determine the molecular basis for the loss of fibronectin receptor function by the FN-1 cells. Unresolved at present is the genetics of the mutation in FN-1. Although MNNG is known to be a point mutagen, the mutation could be either genetic or epigenetic, that is, a defect in a gene product or in gene expression. Preliminary studies indicate that FN-1 reverts spontaneously with a frequency of about 1 in 104, which is borderline between the genetic and epigenetic possibilities (Lewin, 1980). In order to clarify this point further it will be necessary to characterize other BHK cell adhesion variants and carry out complementation analyses. ACKNOWLEDGEMENTS. We are indebted to Mr. Doug McAbee and Dr. Snell for their helpful comments regarding this research. research was supported by a grant from the NIH, CA14609.
Bill This
Cell Biology
178
International
Reports,
Vol. 8, No. 2, February
1984
REFERENCES BRILES, E.B. & HASKEW, N.B. (1982) Isolation a rat hepatoma cell line with altered but normal attachment to fibronectin.
of cloned variants of attachment to collagen, Exp. Cell Res. 138,
436-442.
CHEUNG, E. & JULIANO, R.L. (1983) CHO cell fibronectin coated beads. J. Cell Biol.
aggregation 97, 326a.
induced
by
GRINNELL, F. (1980) The fibroblast receptor for cell-substratum adhesion: Studies on the interaction of baby hamster kidney cells with latex beads coated by cold insoluble globulin (plasma fibronectin). J. Cell Biol. 86, 104-112. GRINNELL, R. (1983) Cell attachment and spreading factors. In: G. Guroff (ed.). Growth and Maturation Factors. pp.267-292. John Wiley and Sons, New York. GRINNELL, F. & FELD, M. (1979) Initial adhesion of blasts in serum-free medium. Possible role fibronectin. Cell c, 117-129. GRINNELL, F. SC HAYS, D.G. (1978) Induction substratum adsorbed ligands directed surface. 116, 275-284. Exp. Cell Res.
of
GRINNELL, F., FELD, M. SCMINTER, D. (1980) Cell nogen and fibrin substrata: Role of cold (plasma fibronectin). Cell 2, 517-525. HARPER, P.A. SC JIJLIANO, R.L. (1981) adhesion of fibroblasts to the Mediation by a high molecular weight J. Cell Biol. 1, 647-653. LEWIN, B. (1980) Gene Expression York pp. 142-188.
Vol.
II.
human fibroof secreted
cell spreading by against cell the adhesion insoluble
to fibriglobulin
Fibronectin-independent extracellular matrix: membrane glycoprotein. John Wiley
and Sons,
New
McABEE, D. & GRINNELL, F. (1983) Fibronectin mediated binding and phagocytosis of polystyrene latex beads by baby hamster kidney cells. J. Cell Biol. In Press. RAFF, N.C. SCDE PETRIS, S. (1973) Movement of lymphocyte surface The fluid nature of the lymphocyte antigens and receptors: significance. immunological Fed. plasma membrane and its Proc. 32, 48-54. SINGER, S.J., ASH, J.F., BOURGNIGNON, L.Y., HEGGENESS, M.H. SC LOUVARD, D. (1978) Transmembrane interactions and the mechanisms of transport of proteins across membranes. J. Supramol. Struct. 9, 373-389. Received: 12th December 1983. Revised version accepted: 31st January 1984.
International
Reports,
Vol. 8, No. 2, February
171
1984
BHK CELL VARIANT WITH DEFECTIVE FIBRONECTIN RECEPTOR FUNCTION Nancy Oppenheimer-Marks, Barbara Border, and Frederick Grinnell Department of Cell Biology University of Texas Health Science Center Dallas, Texas 75235 Running
title:
Fibronectin-receptor
defective
BHK cell
variants
Baby hamster kidney cells were mutagenized with ABSTRACT. N-methyl-N'-nitro-N-nitrosoguanidine and selected to obtain a The cell variant FN-1 was population of non-attaching cells. recloned, and cloned from the non-attaching cell population, tested for cell adhesive interactions using four different assays of fibronectin (pFN) receptor function: cell attachment and spreading on culture dishes and cell binding and phagocytosis of latex beads. On pFN-coated culture dishes, FN-1 cells had decreased attachment compared to parental cells and were unable to as many pFN-bead spread. With pFN-coated beads, only one third binding sites could be detected on FN-1 cells as on the parental cells, and the FN-1 cells were unable to phagocytose the pFNIn other studies, the variant cells were able to coated beads. attach normally and spread partially on substrata coated with polycationic ferritin, concanavalin A, or anti-BHK cell surface antibody. The results suggest that the pFN-receptor function of FN-1 cells is defective. The attachment and spreading of fibroblasts has INTRODUCTION. been shown to occur on a variety of naturally occurring (e.g., fibronectin) or experimental (e.g., lectins) ligand-coated substrata (reviewed in Grinnell, 1983). The meaning of the redundancy in ligands that can promote cell adhesion is unclear. One possibility is that there are independent receptors on the cell surface for each ligand, and that binding between receptor and ligand gives rise to a common signal resulting in cytoskeletal reorganization. Such a model is analagous to that proposed for capping of independent ligands on the cell surface (Raff and DePetris, 1973; Singer -et -', al 1978). One approach to studying this problem has been the use of cell variants with altered adhesive properties. Harper and Juliano (1980), have described a CHO variant that was unable to attach or spread on pFN-coated substrata but attached and spread normally on substrata coated with extracellular matrix, Con A, or poly-L-lysine. Also, Briles and Haskew (1982) described a hepatoma cell variant that attached and spread normally on 0309-l
651/841020171-08/$03.00/O
@ 1984 Academic
Press Inc. (London)
Ltd.
172
Cell Biology
international
Reports,
Vol. 8, No. 2, February
1984
pFN-coated substrata but not on collagen films. One question unresolved by these studies, however, was whether the cell surface receptors were still functional on the cells that failed to attach to the ligand-coated substrata. For instance, Juliano has found that the CHO cell variant incapable of attaching on pFN-coated substrata (see above) binds pFN-coated beads as well as parental cells (Cheung and Juliano, 1983). In order to learn more about the pFN receptor we have begun to study adhesion-deficient BHK cell variants. In these studies, we have used four different assays of pFN-receptor function: cell-substratum attachment, cell-substratum spreading, cell-bead binding, and cell-bead phagocytosis. In addition, the ability of the variant cells to attach and spread on a variety of different ligand-coated substrata has been characterized. The results of these studies are reported herein. MATERIALS AND METHODS. Cells. BHK cells were obtained from Dr. Adrian Chappel (Center for Disease Control, Atlanta, GA) and grown in suspension culture (Grinnell et al., 1980). Exponentially growing cells were harvested bycezrifugation at 800 x g (Sorvall HL-4 rotor) for 4 minutes at room temperature. Cells were washed in Dulbecco's PBS (DPBS) (150 mM NaCl, 3 mM KCl, 6 mM NazHPO,, 1 mM KHzP04, 1 mM CaClz, 0.5 mM MgCl,, pH 7.2) containing 1% bovine serum albumin (BSA). Preparation of FN-1 Cells. BHK cells (107) in 40 ml of minimal essential medium (MEM) were mutagenized with N-methyl-N'-nitro-Nnitrosoguanidine (MNNG) (1 pg/ml) for 4 hours at 37O. The cells were removed from the mutagen by centrifugation for 3 min at 800 x in 20 ml of MEM plus 10% fetal calf serum, and g, resuspended incubated at 37' for 7 days to allow for recovery from the mutagen. Mutagenized BHK cells were subjected to selection by an enrichment procedure to obtain a population of non-attaching cells. On day 1, mutagenized cells (3 x lo7 in 150 ml MEM containing 10% serum) were incubated in Falcon 250 ml tissue attaching culture flasks for 10 min at 37O to remove rapidly were collected, centrifuged, cells. Loosely attached cells resuspended in 100 ml of MEM plus 10% serum and incubated for 2 hours in 250 ml tissue culture flasks at 37O. This procedure was repeated twice and by the end of the third cycle 4.6% of the starting mutagenized cell population had been selected as nonattaching cells. The non-attached cells then were grown in At this point, the suspension culture to obtain 14 x lo7 cells. selection procedure, three 2 hour incubation cycles in tissue Selection was carried out a total culture flasks, was repeated. of 15 times (3 attachment cycles each time) until 95% of the cell population was unable to attach to tissue culture flasks in MEM
Cell Biology
plus
International
Reports,
Vol. 8, No. 2, February
1984
173
10% serum.
After the last selection cycle, the non-attached cells were Cells were diluted in MEM plus 10% cloned by limiting dilution. Aliquots (1 ml) serum to a final concentration of 0.5 cells/ml. of the diluted cells then were incubated in 24 well Falcon tissue Clones were incubator. culture dishes at 37O in a humidified for their ability to grown to high cell density and tested The variant line, FN-1, which interact with pFN-coated substrata. is described in this paper, was recloned. Human plasma fibronectin (pFN) Fibronectin and other materials. was prepared by ammonium sulfate precipitation and ion-exchange The specific activity of chromatography (Grinnell et al., 1980). -the plasma fibronectin preparations used was ZOO-400 units/mg. One unit of activity has been defined as the amount of fibronectin required to promote complete cell spreading in a standardized Radiolabeled 3H-pFN was prepared assay (Grinnell al 1980). -et -** similarly as described previously (Grinnell, 1980). The specific activity of radiolabeled fibronectin resulting from this procedure was 9,000-16,000 cpm/pg. Crystalline bovine serum albumin (BSA) and polycationic Concanavaferritin (PCF) were obtained from Miles Laboratories. from Sigma Chemical Co. lin A (Con A), and MNNG were purchased Polystyrene latex beads with a diameter of 0.76 uM were obtained from Dow Diagnostics. Minimal essential medium, 0.25% trypsin and fetal bovine serum were obtained from GIBCo. solution, Preparation of ligand-coated beads and substrata. Polystyrene latex beads were coated with 'H-pFN as described previously 1980). (Grinnell, The conditions were adjusted to give a final pFN concentration on the beads of 350 ng/cm2 or greater. Just prior to adding the beads to the cells, the bead preparations were sonicated for 5 seconds at 60 watts (Heat Systems Ultrasonics, Inc.) in order to disperse bead aggregates. Falcon #3001 (35 mm) tissue culture dishes were coated with pFN or other ligands as described previously (Grinnell and Hays, 1978) at the ligand concentrations indicated. Dishes coated with ligands were countercoated with DPBS - 1% heat denatured BSA and rinsed with DPBS immediately prior to adding the cells (Grinnell and Feld, 1979). Adhesion and bead binding assays. Cell attachment and cell spreading were measured using standard techniques (Grinnell et al., 1980). To measure bead binding and phagocytosis, cells (2; i?) were mixed with 3H pFN-coated 0.76 uM beads as indicated in 1.0 ml DPBS - 1% BSA. After 20 min at the temperature indicated, the samples were washed, resuspended in 1.0 ml of DPBS - 1% BSA and centrifuged through 2.0 ml of DPBS - 5% BSA. Unbound beads
174
Cell Biology
International
Reports,
Vol. 8, No. 2, February
1984
100
I- 80 E P ” 60 i! s j
40
Y
.-m.
s
FN-1
/
20
0 LA 0
50
loo
Figure
1.
Cell
PCF img/mll
Con A ~pglmll
pFN tpg/ml)
attachment
to various
AMI-EHK
(j.tglmll
substrata
Substrata were coated with the concentrations of the indicated. Coated substrata were incubated with parental cells and the extent of cell attachment measured. Other are in Materials and Methods.
--
8
A .O
PARENT
C -0-o
-PO
ligands or FN-1 details
D
0
0
Figure
2.
Cell
FN-I w..-0 50 1b6 pFN lpglml)
spreading
0
50
100500
Con A fpg/ml)
on various
0
1.0 2.0 0 50 100 200 PCF (mglml) Anti-BHK (pglml)
substrata
Substrata were coated with the concentrations of the ligands indicated. Coated substrata were incubated with parental or FN-1 cells and the extent of cell spreading measured. Other details are in Materials and Methods.
Cell Biology
International
Reports,
Vol. 8, No. 2, February
1984
with cells and whereas cells remained in the supernatant; the sedimented cells were Samples of sedimented. bound beads mixed with 10 ml of Budget Solve (RPI Corp.), and radioactivity was measured in a Nuclear Chicago Mark II scintillation spectroThe remaining sedimented cells were mixed with 1.0 ml photometer. of 0.25% trypsin and incubated for 5 min at 37O to remove surface bound beads but not ingested beads. After these incubations were stopped by the addition of 1.0 ml of fetal calf serum, the cells resuspended in 1.0 ml of DPBS - 1% BSA, and were centrifuged, Beads that were centrifuged through 2 ml of DPBS - 5% BSA. removed by trypsin treatment remained in the supernatant; whereas Samples of the cells and cells with ingested beads sedimented. The sedimented cells were analysed for radioactivity as above. number of total beads bound per cell was calculated as: (cell bead)(cells/ radioactivity) f (radioactivity/ associated The number of ingested beads per cell was calculated incubation). similarly using the cell ingested radioactivity (see McAbee and Grinnell, 1983). RESULTS. BHX cells were mutagenCell Adhesion to Ligand-Coated Substrata. in serum-containing ized with MNNG. Cells that were non-adhesive medium were selected and cloned as described in Materials and The variant clone "FN-1" did not attach to pFN-coated Methods. substrata pre-coated with 5 ug/ml of pFN, which was a concentration sufficient to permit about 90% of the parental cells to As the pFN-concentration was increased, some attach (Figure 1A). attachment of FN-1 cells was observed with 28% of the cells attaching at pFN concentrations of 20 ug/ml or higher. At all pFN concentrations tested, however, FN-1 cells were unable to spread (Figure 2A). When tested on substrata coated with different experimental ligands, FN-1 cells were found to attach to Con A, polycationic ferritin (PCF), and antibodies to BHK cell surfaces (anti-BRK) almost as well as parental cells (Figure lB-D). Spreading of the variant cells occurred on these substrata, but was less than that observed with the parentals (Figure 2B-D). In many cases, the variant cells appeared to extend filipodia but were unable to complete the spreading process. Cell Binding of pFN-Coated Beads. The inability of cells to attach and spread on pFN-coated substrata is not sufficient evidence that pFN receptors are missing or defective (see Introduction). Therefore, experiments also were carried out to determine how well FN-1 cells interacted with pFN-coated beads. Previous studies have demonstrated that the cell-bead binding assay has characteristics typical of a direct ligand-receptor binding assay (Grinnell, 1980; McAbee and Grinnell, 1983). As shown in Table I, the ability of FN-1 cells to bind pFN-coated beads was less than that of the parental cells. On average, the
Cell Biology
176
Table
I.
Binding
International
of pFN-coated
Bead Number
Reports,
Vol. 8, No. 2, February
beads to cells
at 40a
Beads Bound Per Cellb
Added
Parental
% Binding
FN-1 Cells
Cells
FN-l/Parent
3.1 x 109
47 2 2.1
16.5 2 2.1
35.1
6.25
71.2
27.0
5 1.79
37.9
49.7
2 1.16
34.1
x 10'
1.25 x lOlo
145.8
5 18.7
1984
Parental or FN-1 cells were incubated with 3H-pFN-coated a) beads for 20 min at 4O. At the end of the incubations, the radioactivity associated with the cells was measured, and bead binding calculated. Other details are in Materials and Methods.
b)
Mean and S.D.
Table
II.
of duplicate
Phagocytosis
experiments.
of pFN-coated
Beads
beads to cells
Beads Ingested
Cell
Total
Type
Bound/Cellb
Parent
158 + 20.8
50.0
FN-1
40.7
2.1 k .21
2 .49
per Cellb
a>
k 1.16
at 370a
% Ingestion 31.6 5.1
Parental or FN-1 cells were incubated with 2.1 x 10' 3H-pFNcoated beads for 20 min at 37O. At the end of the incubations, the total radioactivity associated with the cells and the ingested radioactivity were measured, and bead binding and ingestion were calculated. Other details are in Materials and Methods.
b)
Mean and S.D.
of duplicate
experiments
Cell Biology
international
variants appeared the parentals.
Reports,
to have
Vol. 8, No. 2, February
35% of the
receptor
1984
binding
capacity
of
The ability of FN-1 cells to phagocytose pFN-coated beads As shown in Table II, total bead binding at 37O also was tested. (surface bound plus ingested) by the variant cells was about 25% Of the beads bound to the parental of that of the parental cells. 31.6% were inside the cells; whereas, only 5.1% of the cells, It can be total bound beads were inside the variant cells. concluded, therefore, that the variants are deficient not only in spreading on pFN-coated substrata, but also, in phagocytosis of pFN-coated beads. The purpose of the studies described in this paper DISCUSSION. was to use adhesion-defective BHE cells to learn more about the involved in cell attachment and spreading. The adhesion sites variant described, FN-1, was shown to have decreased cell attachment and no cell spreading on pFN-coated substrata. Also, of cell surface-pFN bead binding activity was only one third this is the strongest evidence for a detected. To our knowledge, cell variant with inactive pFN receptors. Also, this is the first instance in which a phagocytosis-deficient cell variant has been described. Previously, the possibility was suggested that cell spreading and phagocytosis involve similar cell responses to different sized substrata (Grinnell, 1980). The inability of FN-1 cells to phagocytose pFN-coated beads provides further evidence that spreading and phagocytosis are closely related. At the present time, studies have been initiated to try to determine the molecular basis for the loss of fibronectin receptor function by the FN-1 cells. Unresolved at present is the genetics of the mutation in FN-1. Although MNNG is known to be a point mutagen, the mutation could be either genetic or epigenetic, that is, a defect in a gene product or in gene expression. Preliminary studies indicate that FN-1 reverts spontaneously with a frequency of about 1 in 104, which is borderline between the genetic and epigenetic possibilities (Lewin, 1980). In order to clarify this point further it will be necessary to characterize other BHK cell adhesion variants and carry out complementation analyses. ACKNOWLEDGEMENTS. We are indebted to Mr. Doug McAbee and Dr. Snell for their helpful comments regarding this research. research was supported by a grant from the NIH, CA14609.
Bill This
Cell Biology
178
International
Reports,
Vol. 8, No. 2, February
1984
REFERENCES BRILES, E.B. & HASKEW, N.B. (1982) Isolation a rat hepatoma cell line with altered but normal attachment to fibronectin.
of cloned variants of attachment to collagen, Exp. Cell Res. 138,
436-442.
CHEUNG, E. & JULIANO, R.L. (1983) CHO cell fibronectin coated beads. J. Cell Biol.
aggregation 97, 326a.
induced
by
GRINNELL, F. (1980) The fibroblast receptor for cell-substratum adhesion: Studies on the interaction of baby hamster kidney cells with latex beads coated by cold insoluble globulin (plasma fibronectin). J. Cell Biol. 86, 104-112. GRINNELL, R. (1983) Cell attachment and spreading factors. In: G. Guroff (ed.). Growth and Maturation Factors. pp.267-292. John Wiley and Sons, New York. GRINNELL, F. & FELD, M. (1979) Initial adhesion of blasts in serum-free medium. Possible role fibronectin. Cell c, 117-129. GRINNELL, F. SC HAYS, D.G. (1978) Induction substratum adsorbed ligands directed surface. 116, 275-284. Exp. Cell Res.
of
GRINNELL, F., FELD, M. SCMINTER, D. (1980) Cell nogen and fibrin substrata: Role of cold (plasma fibronectin). Cell 2, 517-525. HARPER, P.A. SC JIJLIANO, R.L. (1981) adhesion of fibroblasts to the Mediation by a high molecular weight J. Cell Biol. 1, 647-653. LEWIN, B. (1980) Gene Expression York pp. 142-188.
Vol.
II.
human fibroof secreted
cell spreading by against cell the adhesion insoluble
to fibriglobulin
Fibronectin-independent extracellular matrix: membrane glycoprotein. John Wiley
and Sons,
New
McABEE, D. & GRINNELL, F. (1983) Fibronectin mediated binding and phagocytosis of polystyrene latex beads by baby hamster kidney cells. J. Cell Biol. In Press. RAFF, N.C. SCDE PETRIS, S. (1973) Movement of lymphocyte surface The fluid nature of the lymphocyte antigens and receptors: significance. immunological Fed. plasma membrane and its Proc. 32, 48-54. SINGER, S.J., ASH, J.F., BOURGNIGNON, L.Y., HEGGENESS, M.H. SC LOUVARD, D. (1978) Transmembrane interactions and the mechanisms of transport of proteins across membranes. J. Supramol. Struct. 9, 373-389. Received: 12th December 1983. Revised version accepted: 31st January 1984.